10:25 < Oy> happy new year :) 10:31 <@heinrich5991> happy new year :) 10:32 < Teeworlds> [teeworlds] oy pushed 3 new commits to master: https://git.io/fhqGs 10:32 < Teeworlds> teeworlds/master 51b2b5c Speedy Consoles: Fix audio panning 10:32 < Teeworlds> teeworlds/master f06a5eb Speedy Consoles: Add square-law panning 10:32 < Teeworlds> teeworlds/master d6cb666 oy: Merge pull request #1940 from Speedy-Consoles/master... 10:43 < bridge> [teeworlds] happy new year guys 12:30 <@Dune> hey ho 12:30 <@Dune> happy new year, Oy 12:30 <@Dune> would you mind clearing up why that wouldn't work? https://github.com/teeworlds/teeworlds/issues/1941#issuecomment-450818339 12:33 < Oy> well, the client has no clue which commands will be available 12:33 < Oy> and doesn't know when loading the available commands is finished 12:33 <@Dune> oh 12:34 <@Dune> so you can't delete commands at the end 12:34 < Oy> there's no end msg by the server 12:35 < Oy> already found the problem, will fix it later, so it's as fast as 0.6 12:35 <@Dune> oh, thanks :) 13:03 <@Dune> any clue how to make gametype icons support both types? check their size and only allow square and 1:3? 13:03 < bridge> [teeworlds] Both types? 13:04 <@Dune> https://github.com/teeworlds/teeworlds/issues/1903 13:09 <@Dune> Transifex is seeing progress, by the way 13:09 <@Dune> We didn't add more strings recently, right 13:11 <@Dune> all major languages are >90% except turkish and chinese 13:14 <@heinrich5991> Learath2: ^? :) 13:16 < Oy> think there are some new ones 13:16 < Oy> also the game messages (join capture) changes because of the id right? 13:17 < Oy> about the gametype icons: yeah check the size 13:19 <@Dune_> those were all in 0.7.1? 13:21 < Oy> yeah 13:21 <@Dune_> so should be updated already 13:22 < Oy> but the change was after i think 13:22 < Oy> checking 13:23 <@Dune_> ho, russian still has "K" and "D" strings 13:23 <@Dune_> those aren't used anymore afaik 13:23 < Oy> https://github.com/teeworlds/teeworlds/commit/88622ec94d96b82920e79366b9c49b0e8b44f3be was 0.7.2 13:23 <@Dune_> ah i see, thx 13:26 <@heinrich5991> btw, for translation it's usually best if we don't have strings like str_format(Localize("%s has joined the %s team"), …, Team ? Localize("blue") : Localize("red")), but instead str_format(Team ? Localize("%s has joined the blue team") : Localize("%s has joined the red team"), …) 13:26 <@heinrich5991> I saw a few instances of that in the source code, might be better with the other version going forward 13:27 <@heinrich5991> (grammar works weirdly in many languages, and it's more intuitive to translate the whole string) 13:28 <@Dune_> ah yeah, that is true 13:28 <@Dune_> usually having several %s in a string is bad because languages tend to enjoy swapping things 13:29 <@heinrich5991> ah yes, that's also I wanted to fix 13:29 <@Dune_> (well if you fix that you fix the former issue) 13:29 <@heinrich5991> not really, you still have to translate five versions of "blue" for each language, depending on the context. it's easier to do it in one place for one string 13:30 <@Dune_> ah right 13:30 <@heinrich5991> e.g. for "Join blue" you need a different string (not sure if we use "Join %s" there) 13:30 <@Dune_> w edont 13:30 <@heinrich5991> but you get the point :) 13:30 <@Dune_> :) 13:33 <@Dune_> exporting the last strings 14:00 < bridge> [teeworlds] Serbian has K D strings too 14:00 <@Dune_> it does? 14:00 <@Dune_> huh 14:01 <@Dune_> that is strange 14:04 <@Dune_> probably need to cleanup something before running script, will check when I'm back 14:04 <@Dune_> thanks for the report 15:48 < Teeworlds> [teeworlds] oy pushed 2 new commits to master: https://git.io/fhqVj 15:48 < Teeworlds> teeworlds/master 0027eff oy: fixed line endings 15:48 < Teeworlds> teeworlds/master 7b21682 oy: speeded up rcon command update process. closes #1941 15:51 < bridge> [teeworlds] @LordSk ^ :) 15:52 < bridge> [teeworlds] Nice! 15:55 < Oy> twice as fast as it used to be in 0.6, but should be good 15:58 < bridge> [teeworlds] yay 16:40 <@Dune_> https://github.com/teeworlds/teeworlds/blob/7b21682fec2c89b395b4eafe7ff9ab98707a6043/src/game/client/components/scoreboard.cpp#L322-L327 16:40 <@Dune_> no shit K is still in the strings 16:41 <@Dune_> lesson: don't commit at 9am 16:58 < Oy> :P 17:47 < bridge> [teeworlds] damn im getting banned a lot lately 😄 18:30 < Oy> github down :( 19:27 < Teeworlds> [teeworlds] oy pushed 1 new commit to master: https://git.io/fhq5R 19:27 < Teeworlds> teeworlds/master 2d2fe8b oy: reverted edit boxes to 0.6 behavior. closes #1937 19:46 < Teeworlds> [teeworlds] oy closed pull request #1730: Add CMake (and a couple of other fixes) (master...pr_cmake) https://git.io/fpu1t 19:53 < Teeworlds> [teeworlds] oy pushed 1 new commit to master: https://git.io/fhqFU 19:53 < Teeworlds> teeworlds/master c009476 oy: fixed sdl2 & freetype lib paths. changed macos bundleidentifier for cmake to match previous one 20:04 < Teeworlds> [teeworlds] oy pushed 1 new commit to master: https://git.io/fhqFb 20:04 < Teeworlds> teeworlds/master b6c2e2f oy: always show scoreboard within server info. don't hide bots in case no filter is selected. closes #1943 20:19 <@Dune> oh, so cmake is in 21:05 < bridge> [teeworlds] I have downloaded latest updated, with cmake. I did `git submodule update --init --recursive`, to get maps and stuff. Also I used `python3 scripts/download.py --arch=x86_64 --conf=release sdl freetype` to download libraries and includes. 21:05 < bridge> [teeworlds] CMake says: `You must install Freetype/SDL2 to compile the Teeworlds client`. 21:05 < bridge> [teeworlds] 21:05 < bridge> [teeworlds] Looks like CMake can't find libraries 21:11 <@heinrich5991> getting dependencies automatically is not done for cmake yet 21:15 < bridge> [teeworlds] Don't understand? but files are in correct places 21:15 < bridge> [teeworlds] after all 21:15 < bridge> [teeworlds] should it work ? 21:15 < bridge> [teeworlds] or cmake looking for files in different paths 21:18 <@heinrich5991> different paths 21:27 < bridge> [teeworlds] Okay, I fixed that 21:33 < Teeworlds> [teeworlds] savander opened pull request #1944: fix paths to libs in CMakeLists.txt (master...patch-1) https://git.io/fhmes 21:35 < Oy> Savander that would probably make cross-compiling pretty hard 21:35 < bridge> [teeworlds] why? 21:36 < Oy> heinrich5991: can you build cross-compile all plattforms with cmake? 21:36 <@heinrich5991> yes 21:36 < bridge> [teeworlds] linux or mac, have different path to lib? 21:36 <@heinrich5991> but not with that PR merged, probably 21:36 < bridge> [teeworlds] okay, so just close it up. 😉 21:37 < Oy> Savander you have to check if you'Re building on windows, then u can use these paths ;) 21:37 < Teeworlds> [teeworlds] savander closed pull request #1944: fix paths to libs in CMakeLists.txt (master...patch-1) https://git.io/fhmes 21:38 < bridge> [teeworlds] so linux, and mac have different paths? 21:38 < Oy> no 21:38 < Oy> well yeah 21:38 < Oy> for cross-compiling you need different paths for the libs u want to build for 21:38 < Oy> *with 21:39 < bridge> [teeworlds] i will not mess with it :PP then 21:39 < Oy> when you'Re on linux and you want to build the windows binary you have to use the windows libs, same for mac - doesn't work if it just looks in 1 path then 21:41 < bridge> [teeworlds] shouldn't download script, download libs to other///lib/ 21:41 < bridge> [teeworlds] ? 21:41 < bridge> [teeworlds] then? 21:41 < Oy> then actual working directories wouldn't work/update anymore 21:42 < bridge> [teeworlds] how it was before, i remember that it was like: "other/sdl/windows/??" 21:42 < bridge> [teeworlds] and "other/sdl/include" 21:43 < Oy> other/sdl/lib/32 other/sdl/include - like when u unpack the official sdl archive u download 21:44 < Oy> other/sdl/lib/x86 other/sdl/include - like when u unpack the official sdl archive u download 21:45 < bridge> [teeworlds] hmm, im confused then. Maybe I'll leave it 21:47 < Oy> well, with the download script we could actually use the cmake path and then just rely on people having internet access when building 21:51 < bridge> [teeworlds] I see, bam downloads automatically SDL nad Freetype 21:52 < bridge> [teeworlds] if doesn't exists 21:53 < bridge> [teeworlds] Have you tried to build tw with bam after that update? with auto-downloaded libs ? 21:54 < bridge> [teeworlds] cause i have linking error 21:56 < bridge> [teeworlds] nevermind. Forgot 22:04 < bridge> [teeworlds] or no, why i said that. 😄 I'm a bit nervous today, sorry. 22:04 < bridge> [teeworlds] So, The there's at least one thing. 22:05 < bridge> [teeworlds] bam compiled with CL, using compiler=cl on teeworlds compiling 22:05 < bridge> [teeworlds] it can't download libs because of : `duownload.py: error: argument --arch: invalid choice: 'unknown'` 22:05 < bridge> [teeworlds] also, i can't specify arch because: Cross compiling is unsupported on Windows. 22:06 < bridge> [teeworlds] the "unknow" arch, is set up automatically 22:08 < bridge> [teeworlds] Gettimng a bunch of automated mails about fail compil because of immintrin.h 22:09 < bridge> [teeworlds] Should disable the code or something at least 22:10 < bridge> [teeworlds] with default compiler, "gcc", arch=x86. It downloads libs but there's linking error: https://pastebin.com/29U9nWBy 22:10 < bridge> [teeworlds] If anyone else can test, it would be great 😉 22:11 < bridge> [teeworlds] Pure teeworlds-master clone 22:12 < Teeworlds> [teeworlds] heinrich5991 opened pull request #1945: Remove unportable and `_mm_pause` function (master...pr_remove_immintr) https://git.io/fhmfx 22:13 < Oy> which platforms report these problems? they probably can't run the client anyway 22:14 <@heinrich5991> e.g. arm 22:15 < Oy> yeah raspberry pi can't run the client 22:15 <@heinrich5991> huh, can't it? 22:15 < Oy> no 22:16 < Oy> have to update to opengl es 22:16 < Oy> then it can 22:16 < Oy> otherwise u just get 1-2 frames a second 22:17 <@heinrich5991> ah 22:18 <@heinrich5991> the thing that this instruction does is tell the CPU to relax its execution of this hyperthread 22:18 <@heinrich5991> that means it benefits the case where something else wants to do something with the second hyperthread on the same CPU 22:18 <@heinrich5991> otherwise it's still a spinloop. I think it's not worth the pain to keep 22:18 <@heinrich5991> we can of course guard it behind #if (intel processor) 22:20 < Oy> yeah use it for x86/x64 arch only 22:36 < bridge> [teeworlds] +1 22:38 < Teeworlds> [teeworlds] heinrich5991 closed pull request #1945: Remove unportable and `_mm_pause` function (master...pr_remove_immintr) https://git.io/fhmfx 22:38 <@heinrich5991> if you implement that, please also move it to src/base/system.c 22:43 < Oy> yeah, best place for it 22:55 < Oy> guess intel itanium doesn't support it too - not sure 23:00 < Teeworlds> [teeworlds] oy pushed 1 new commit to master: https://git.io/fhmkQ 23:00 < Teeworlds> teeworlds/master 87fa774 oy: made the mm_pause instruction only available for intel32 and amd64 architectures. closes #1882 23:04 < bridge> [teeworlds] Thanks! 23:20 < Teeworlds> [teeworlds] oy pushed 1 new commit to master: https://git.io/fhmLB 23:20 < Teeworlds> teeworlds/master 5bd7758 oy: changed path structure for sdl & freetype libs on windows so cmake can use the ones provided by the bam downloader